Image Desaturation for SDO/AIA Using Deep Learning

Yu, Xuexin; Xu, Long; Yan, Yihua

doi:10.1007/s11207-021-01808-2

Cited by 9 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With significant success of deep learning in image inpainting, two learning-based approaches, Mask-Pix2Pix (Zhao et al 2019) and PCGAN (Yu et al 2021), were proposed to desaturate solar images in our previous efforts. They differ from DESAT (Schwartz et al 2015) and SE-DESAT (Guastavino et al 2019) in three aspects.…”

Section: Introductionmentioning

confidence: 99%

“…Once a standard convolution slides to the boundary of saturated region, invalid pixels would participate in convolution, resulting in deviation of convolution, e.g, Mask-Pix2Pix (Zhao et al 2019). To overcome this problem, partial convolution (PC) (Liu et al 2018) was employed to replace standard convolution in our previous effort (Yu et al 2021), which excludes saturated pixels from block-wise convolution and compensates deviation of PC to approach normal convolution as far as possible.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Image Desaturation for SDO/AIA Using Mixed Convolution Network

Yu¹,

Xu²,

Ren³

et al. 2022

Res. Astron. Astrophys.

View full text Add to dashboard Cite

The Atmospheric Imaging Assembly (AIA) onboard the Solar Dynamics Observatory (SDO) provides full-disk solar images with high temporal cadence and spatial resolution over seven extreme ultraviolet (EUV) wavebands. However, as violent solar flares happen, images captured in EUV wavebands may have saturation in active regions, resulting in information loss. In this paper, we propose a deep learning model to restore the lost signal in saturated regions by referring to both unsaturated/normal regions within a solar image and statistical probability model of massive normal solar images. The proposed model, namely mixed convolution network (MCNet), is established over conditional generative adversarial network (GAN) and the combination of partial convolution (PC) and validness migratable convolution (VMC). These two convolutions were originally proposed for image inpainting. In addition, they are implemented only on unsaturated/valid pixels, followed by certain compensation to compensate the deviation of PC/VMC relative to normal convolution. Experimental results demonstrate that the proposed MCNet achieves favorable desaturated results for solar images and outperforms the state-of-the-art methods both quantitatively and qualitatively.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Image Desaturation for SDO/AIA Using Mixed Convolution Network

Yu¹,

Xu²,

Ren³

et al. 2022

Res. Astron. Astrophys.

View full text Add to dashboard Cite

show abstract

“…Our method is much simpler than alternative image completion techniques based on artificial intelligence/deep learning. Such methods have been successfully applied to the restoration of images of global positioning system (GPS) measurements of the ionosphere (Chen et al 2019;Pan et al 2020), as well as solar images corrupted by flares (Yu et al 2021). Deep learning techniques usually rely on training of a set of artificial neural networks using reference data before the networks can be used to fill the gaps in real observations.…”

Section: Discussionmentioning

confidence: 99%

Reconstruction of photospheric velocity fields from highly corrupted data

Rempel,

Chertovskih,

Davletshina

et al. 2022

Preprint

View full text Add to dashboard Cite

The analysis of the photospheric velocity field is essential for understanding plasma turbulence in the solar surface, which may be responsible for driving processes such as magnetic reconnection, flares, wave propagation, particle acceleration, and coronal heating. Currently, the only available methods to estimate velocities at the solar photosphere transverse to an observer's line of sight infer flows from differences in image structure in successive observations. Due to data noise, algorithms such as local correlation tracking (LCT) may lead to a vector field with wide gaps where no velocity vectors are provided. In this letter, a novel method for image inpainting of highly corrupted data is proposed and applied to the restoration of horizontal velocity fields in the solar photosphere. The restored velocity field preserves all the vector field components present in the original field. The method shows robustness when applied to both simulated and observational data.

show abstract

“…Our method is much simpler than alternative image completion techniques based on artificial intelligence/deep learning. Such methods have been successfully applied to the restoration of images of global positioning system (GPS) measurements of the ionosphere (Chen et al 2019;Pan et al 2020), as well as solar images corrupted by flares (Yu et al 2021). Deep-learning techniques usually rely on training of a set of artificial neural networks using reference data before the networks can be used to fill the gaps in real observations.…”

Section: Discussionmentioning

confidence: 99%

Reconstruction of Photospheric Velocity Fields from Highly Corrupted Data

Rempel

Chertovskih

Davletshina

et al. 2022

ApJ

View full text Add to dashboard Cite

The analysis of the photospheric velocity field is essential for understanding plasma turbulence in the solar surface, which may be responsible for driving processes such as magnetic reconnection, flares, wave propagation, particle acceleration, and coronal heating. Currently, the only available methods to estimate velocities at the solar photosphere transverse to an observer’s line of sight infer flows from differences in image structure in successive observations. Due to data noise, algorithms such as local correlation tracking may lead to a vector field with wide gaps where no velocity vectors are provided. In this paper, a novel method for image inpainting of highly corrupted data is proposed and applied to the restoration of horizontal velocity fields in the solar photosphere. The restored velocity field preserves all the vector field components present in the original field. The method shows robustness when applied to both simulated and observational data.

show abstract

Image Desaturation for SDO/AIA Using Deep Learning

Cited by 9 publications

References 9 publications

Image Desaturation for SDO/AIA Using Mixed Convolution Network

Image Desaturation for SDO/AIA Using Mixed Convolution Network

Reconstruction of photospheric velocity fields from highly corrupted data

Reconstruction of Photospheric Velocity Fields from Highly Corrupted Data

Contact Info

Product

Resources

About